Psychiatry and Clinical Neurosciences 2015; 69: 360–368
doi:10.1111/pcn.12255
Regular Article
White matter abnormalities in major depressive disorder with melancholic and atypical features: A diffusion tensor imaging study Miho Ota, MD, PhD,1* Takamasa Noda, MD,3 Noriko Sato, MD, PhD,4 Kotaro Hattori, MD, PhD,1 Hiroaki Hori, MD, PhD,1 Daimei Sasayama, MD,1 Toshiya Teraishi, MD, PhD,1 Anna Nagashima, MA,1 Satoko Obu, MA,1 Teruhiko Higuchi, MD, PhD2 and Hiroshi Kunugi, MD, PhD1 1
Department of Mental Disorder Research, National Institute of Neuroscience, 2National Center of Neurology and Psychiatry, Departments of 3Psychiatry and 4Radiology, National Center Hospital of Neurology and Psychiatry, Tokyo, Japan
Aim: The DSM-IV recognizes some subtypes of major depressive disorder (MDD). It is known that the effectiveness of antidepressants differs among the MDD subtypes, and thus the differentiation of the subtypes is important. However, little is known as to structural brain changes in MDD with atypical features (aMDD) in comparison with MDD with melancholic features (mMDD), which prompted us to examine possible differences in white matter integrity assessed with diffusion tensor imaging (DTI) between these two subtypes. Methods: Subjects were 21 patients with mMDD, 24 with aMDD, and 37 age- and sex-matched healthy volunteers whose DTI data were obtained by 1.5 tesla magnetic resonance imaging. We compared fractional anisotropy and mean diffusivity value derived from DTI data on a voxel-by-voxel basis among the two diagnostic groups and healthy subjects.
AJOR DEPRESSIVE DISORDER (MDD) is a common psychiatric disorder with a lifetime prevalence reported to range from 8% to 12% worldwide.1 The DSM-IV2 defines subtypes of MDD accord-
M
*Correspondence: Miho Ota, MD, PhD, Department of Mental Disorder Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, 4-1-1, Ogawa-Higashi, Kodaira, Tokyo 187-8502, Japan. Email: [email protected] Received 29 January 2014; revised 3 November 2014; accepted 6 November 2014.
360
Results: There were significant decreases of fractional anisotropy and increases of mean diffusivity in patients with MDD compared with healthy subjects in the corpus callosum, inferior fronto-occipital fasciculus, and left superior longitudinal fasciculus. However, we detected no significant difference in any brain region between mMDD and aMDD. Conclusion: Our results suggest that patients with MDD had reduced white matter integrity in some regions; however, there was no major difference between aMDD and mMDD. Key words: atypical features, diffusion tensor imaging, major depressive disorder, melancholic features, tract-based spatial statistics.
ing to the variable clinical presentations. About 30% of unipolar depressive outpatients may meet the DSM-IV criteria for atypical depression.3 Atypical depression is characterized by mood reactivity, increased appetite (or weight gain), hypersomnia, leaden paralysis, and interpersonal rejection sensitivity. Furthermore, patients with atypical depression, as opposed to those with other forms of depression, are more likely to be female and to have a younger age of onset, a longer duration of illness, and a greater likelihood of irritability and psychomotor retardation.4 As
© 2014 The Authors Psychiatry and Clinical Neurosciences © 2014 Japanese Society of Psychiatry and Neurology
Psychiatry and Clinical Neurosciences 2015; 69: 360–368
for the biological underpinnings, atypical depression appears to be associated with abnormally decreased HPA axis function and hypocortisolemia, as well as chronic fatigue syndrome and post-traumatic stress disorder.5,6 The concept of atypical depression was previously recognized as a subgroup of depressed patients who were more responsive to monoamine oxidase inhibitors than tricyclic antidepressants.7 The atypical depression subtype was formally recognized as an ‘episode specifier’ in the DSM-IV. Atypical depression is also common in individuals with bipolar disorder or dysthymic disorder, and thus these disorders could specify the atypical features. In the DSM-V, while the definition of atypical features has not changed, the depressive disorders and bipolar disorder are treated in separate chapters. Previous studies have reported structural brain changes in patients with MDD, particularly in the hippocampus.8–12 Studies using diffusion tensor imaging (DTI) have shown a variety of abnormalities in white matter fibers, such as those of the left superior longitudinal fasciculus,13–15 left inferior front-occipital fasciculus to posterior thalamic radiation,13,16 unilateral or lateral parahippocampus,13,17 bilateral internal capsule and external capsule,13,18 and cingulum to posterior corpus callosum.16,17,19 These wide-ranging findings may have arisen from the heterogeneity of the disorder, as there is some evidence that different MDD subtypes have different behavioral and neurochemical presentations.20,21 Some studies have focused on the relation between structural or functional changes of the brain and MDD subtypes. Greenberg et al. examined the volume differences in hippocampi between MDD patients with atypical features (aMDD) and those with melancholic features (mMDD),12 and another study measured hippocampal volume between mMDD and MDD without melancholic features.8 However, these studies found no significant difference between the subtypes. One DTI study also showed no significant difference in fractional anisotropy (FA) between mMDD and MDD without melancholic features.13 However, this study did not focus on aMDD and was limited by the small sample size (11 mMDD and 18 non-mMDD patients). The present study was designed to examine whether there are any differences in white matter integrity between aMDD and mMDD by using DTI in a relatively large sample size, which was expected to elucidate the structural basis for these subtypes.
WM abnormalities in MDD subtypes 361
METHODS Subjects From September 2009 to September 2013, 170 patients (139 outpatients and 31 inpatients) whose chief complaints were depressed symptoms underwent brain magnetic resonance imaging (MRI) in our institution. All the inpatients were admitted to a psychiatric ward of the National Center of Neurology and Psychiatry Hospital, Japan. Among them, 45 patients with MDD (aMDD = 24 subjects) who fulfilled the DSM-IV criteria2 for aMDD or mMDD were involved in this study. Thirty-seven age- and sexmatched healthy volunteers were recruited from the community through local magazine advertisements and our website announcement. Research psychiatrists with board certification (M.O., H.H., D.S., and T.T.) made the diagnosis and rated depression severity by the 21-item version of the Hamilton Depression Rating scale (HAM-D) for their depressive symptoms.22 Daily doses of antidepressant were converted to imipramine equivalents using published guidelines.23 Healthy volunteers were interviewed for enrollment by a research psychiatrist using the Japanese version of the Mini-International Neuropsychiatric Interview.24,25 Participants were excluded if they had a prior medical history of central nervous system disease or severe head injury, or if they met the criteria for substance abuse or dependence. Those individuals who demonstrated no history of psychiatric illness or contact with psychiatric services were enrolled as controls. After the study was explained to the subjects, written informed consent was obtained for participation in the study from every subject. This study was approved by the ethics committee of the National Center of Neurology and Psychiatry, Japan.
MRI data acquisition and processing MRI was performed on a Magnetom Symphony 1.5tesla (Siemens, Erlangen, Germany). Diffusion was measured along 12 non-collinear directions with the use of a diffusion-weighted factor b in each direction for 1000 s/mm2, and one image was acquired without using a diffusion gradient. The DTI examination took approximately 6 min. In addition to DTI, conventional axial T2-weighted images (echo time [TE]/repetition time [TR]: 95/3500 ms; flip angle: 150°; slice thickness: 5 mm; intersection gap: 1.75 mm; matrix: 448 × 512; field of view [FOV]:
© 2014 The Authors Psychiatry and Clinical Neurosciences © 2014 Japanese Society of Psychiatry and Neurology
362 M. Ota et al.
Psychiatry and Clinical Neurosciences 2015; 69: 360–368
210 × 240 mm2; acquisitions: 1) and fluid attenuation inversion recovery images in the axial plane (TE/ TR: 101/8800 ms; flip angle: 150°; slice thickness: 3 mm; intersection gap: 1.75 mm; matrix: 448 × 512; FOV: 210 × 240 mm2; acquisition: 1) were acquired to exclude cerebrovascular disease. On conventional MRI, no abnormal findings were detected in the brain in any subject included in the study. DTI metrics were calculated by using Tract-Based Spatial Statistics (TBSS) analysis.26 The TBSS algorithm is available in the Functional Magnetic Resonance Imaging of the Brain (FMRIB) Software Library (FSL; version 4.1).27 An FA threshold of 0.20 or higher was set to exclude peripheral tracts.
In addition to the voxel-wise analysis, we examined skeletonized FA and MD data in a regions-ofinterest (ROI) manner for specific white matter tracts derived from the ‘JHU-WhiteMatter-Labels1mm.nii.gz’ regarded as the anatomically standard image atlas in FSL. We evaluated the differences among the two diagnostic groups and healthy subjects using ANCOVA controlling for age, sex, and education, followed by Bonferroni-corrected pairwise comparisons as post-hoc tests using SPSS. A twotailed P-value < 0.05 was regarded as statistically significant.
Statistical analyses We first evaluated the differences in demographic and clinical characteristics among the two diagnostic groups and healthy subjects using ANOVA and between the two patient groups using a two-sample t-test. These tests were performed with SPSS 20 (SPSS Japan, Tokyo, Japan). Skeletonized FA and mean diffusivity (MD) data were analyzed to assess the difference in FA and MD values among the two diagnostic groups and healthy subjects, controlling for age, sex, and education using the Threshold-Free Cluster Enhancement option in FSL’s Randomize with 10 000 permutations. The script uses a permutation-based statistical inference that does not rely on Gaussian distribution of voxels, and runs without having to define an initial clusterforming threshold or carry out a large amount of data smoothing.28,29 The level of statistical significance was set at P < 0.05 with family-wise error (FWE) rate correction for multiple comparisons.
RESULTS Demographic and clinical characteristics of the subjects are shown in Table 1. There was no statistically significant difference in age, sex, or education years across the three groups. There was no significant difference in dose of antidepressant or age of onset between the two patient groups, however the HAM-D score in the mMDD group was significantly higher than that in subjects with aMDD. There were significant decreases of FA in the left superior longitudinal fasciculus (Fig. 1a), left frontooccipital fasciculus (Fig. 1b), and corpus callosum (Fig. 1a,c) in the aMDD patients compared with the healthy subjects (Fig. 1a,c). Significant decreases in FA values were observed diffusely in the mMDD patients compared with the healthy subjects, and the regions exhibiting the FA decreases were similar to those in the subjects with aMDD (Fig. 1d–f). When we compared the FA values between aMDD and mMDD patients, we found no significant difference between the two groups in any region. As for MD, there were significant increases in the left superior
Table 1. Demographic and clinical characteristics of the study sample (mean ± standard deviation)
Male/Female Age (years) Education (years) Age of onset (years) Antidepressant medication (mg)† HAM-D
MDD patients with melancholic features
MDD patients with atypical features
Healthy volunteers
11/10 42.3 ± 10.0 14.7 ± 1.9 33.0 ± 10.8 157.3 ± 186.3 22.0 ± 7.4
11/13 38.3 ± 10.3 14.7 ± 2.2 30.9 ± 9.6 128.7 ± 136.2 14.9 ± 5.2
16/21 40.0 ± 10.0 14.6 ± 2.1
†
Imipramine equivalent dose. HAM-D, Hamilton Rating Scale for Depression; MDD, major depressive disorder.
© 2014 The Authors Psychiatry and Clinical Neurosciences © 2014 Japanese Society of Psychiatry and Neurology
P-value 0.74 0.44 0.93 0.29 0.56
doi:10.1111/pcn.12255
Regular Article
White matter abnormalities in major depressive disorder with melancholic and atypical features: A diffusion tensor imaging study Miho Ota, MD, PhD,1* Takamasa Noda, MD,3 Noriko Sato, MD, PhD,4 Kotaro Hattori, MD, PhD,1 Hiroaki Hori, MD, PhD,1 Daimei Sasayama, MD,1 Toshiya Teraishi, MD, PhD,1 Anna Nagashima, MA,1 Satoko Obu, MA,1 Teruhiko Higuchi, MD, PhD2 and Hiroshi Kunugi, MD, PhD1 1
Department of Mental Disorder Research, National Institute of Neuroscience, 2National Center of Neurology and Psychiatry, Departments of 3Psychiatry and 4Radiology, National Center Hospital of Neurology and Psychiatry, Tokyo, Japan
Aim: The DSM-IV recognizes some subtypes of major depressive disorder (MDD). It is known that the effectiveness of antidepressants differs among the MDD subtypes, and thus the differentiation of the subtypes is important. However, little is known as to structural brain changes in MDD with atypical features (aMDD) in comparison with MDD with melancholic features (mMDD), which prompted us to examine possible differences in white matter integrity assessed with diffusion tensor imaging (DTI) between these two subtypes. Methods: Subjects were 21 patients with mMDD, 24 with aMDD, and 37 age- and sex-matched healthy volunteers whose DTI data were obtained by 1.5 tesla magnetic resonance imaging. We compared fractional anisotropy and mean diffusivity value derived from DTI data on a voxel-by-voxel basis among the two diagnostic groups and healthy subjects.
AJOR DEPRESSIVE DISORDER (MDD) is a common psychiatric disorder with a lifetime prevalence reported to range from 8% to 12% worldwide.1 The DSM-IV2 defines subtypes of MDD accord-
M
*Correspondence: Miho Ota, MD, PhD, Department of Mental Disorder Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, 4-1-1, Ogawa-Higashi, Kodaira, Tokyo 187-8502, Japan. Email: [email protected] Received 29 January 2014; revised 3 November 2014; accepted 6 November 2014.
360
Results: There were significant decreases of fractional anisotropy and increases of mean diffusivity in patients with MDD compared with healthy subjects in the corpus callosum, inferior fronto-occipital fasciculus, and left superior longitudinal fasciculus. However, we detected no significant difference in any brain region between mMDD and aMDD. Conclusion: Our results suggest that patients with MDD had reduced white matter integrity in some regions; however, there was no major difference between aMDD and mMDD. Key words: atypical features, diffusion tensor imaging, major depressive disorder, melancholic features, tract-based spatial statistics.
ing to the variable clinical presentations. About 30% of unipolar depressive outpatients may meet the DSM-IV criteria for atypical depression.3 Atypical depression is characterized by mood reactivity, increased appetite (or weight gain), hypersomnia, leaden paralysis, and interpersonal rejection sensitivity. Furthermore, patients with atypical depression, as opposed to those with other forms of depression, are more likely to be female and to have a younger age of onset, a longer duration of illness, and a greater likelihood of irritability and psychomotor retardation.4 As
© 2014 The Authors Psychiatry and Clinical Neurosciences © 2014 Japanese Society of Psychiatry and Neurology
Psychiatry and Clinical Neurosciences 2015; 69: 360–368
for the biological underpinnings, atypical depression appears to be associated with abnormally decreased HPA axis function and hypocortisolemia, as well as chronic fatigue syndrome and post-traumatic stress disorder.5,6 The concept of atypical depression was previously recognized as a subgroup of depressed patients who were more responsive to monoamine oxidase inhibitors than tricyclic antidepressants.7 The atypical depression subtype was formally recognized as an ‘episode specifier’ in the DSM-IV. Atypical depression is also common in individuals with bipolar disorder or dysthymic disorder, and thus these disorders could specify the atypical features. In the DSM-V, while the definition of atypical features has not changed, the depressive disorders and bipolar disorder are treated in separate chapters. Previous studies have reported structural brain changes in patients with MDD, particularly in the hippocampus.8–12 Studies using diffusion tensor imaging (DTI) have shown a variety of abnormalities in white matter fibers, such as those of the left superior longitudinal fasciculus,13–15 left inferior front-occipital fasciculus to posterior thalamic radiation,13,16 unilateral or lateral parahippocampus,13,17 bilateral internal capsule and external capsule,13,18 and cingulum to posterior corpus callosum.16,17,19 These wide-ranging findings may have arisen from the heterogeneity of the disorder, as there is some evidence that different MDD subtypes have different behavioral and neurochemical presentations.20,21 Some studies have focused on the relation between structural or functional changes of the brain and MDD subtypes. Greenberg et al. examined the volume differences in hippocampi between MDD patients with atypical features (aMDD) and those with melancholic features (mMDD),12 and another study measured hippocampal volume between mMDD and MDD without melancholic features.8 However, these studies found no significant difference between the subtypes. One DTI study also showed no significant difference in fractional anisotropy (FA) between mMDD and MDD without melancholic features.13 However, this study did not focus on aMDD and was limited by the small sample size (11 mMDD and 18 non-mMDD patients). The present study was designed to examine whether there are any differences in white matter integrity between aMDD and mMDD by using DTI in a relatively large sample size, which was expected to elucidate the structural basis for these subtypes.
WM abnormalities in MDD subtypes 361
METHODS Subjects From September 2009 to September 2013, 170 patients (139 outpatients and 31 inpatients) whose chief complaints were depressed symptoms underwent brain magnetic resonance imaging (MRI) in our institution. All the inpatients were admitted to a psychiatric ward of the National Center of Neurology and Psychiatry Hospital, Japan. Among them, 45 patients with MDD (aMDD = 24 subjects) who fulfilled the DSM-IV criteria2 for aMDD or mMDD were involved in this study. Thirty-seven age- and sexmatched healthy volunteers were recruited from the community through local magazine advertisements and our website announcement. Research psychiatrists with board certification (M.O., H.H., D.S., and T.T.) made the diagnosis and rated depression severity by the 21-item version of the Hamilton Depression Rating scale (HAM-D) for their depressive symptoms.22 Daily doses of antidepressant were converted to imipramine equivalents using published guidelines.23 Healthy volunteers were interviewed for enrollment by a research psychiatrist using the Japanese version of the Mini-International Neuropsychiatric Interview.24,25 Participants were excluded if they had a prior medical history of central nervous system disease or severe head injury, or if they met the criteria for substance abuse or dependence. Those individuals who demonstrated no history of psychiatric illness or contact with psychiatric services were enrolled as controls. After the study was explained to the subjects, written informed consent was obtained for participation in the study from every subject. This study was approved by the ethics committee of the National Center of Neurology and Psychiatry, Japan.
MRI data acquisition and processing MRI was performed on a Magnetom Symphony 1.5tesla (Siemens, Erlangen, Germany). Diffusion was measured along 12 non-collinear directions with the use of a diffusion-weighted factor b in each direction for 1000 s/mm2, and one image was acquired without using a diffusion gradient. The DTI examination took approximately 6 min. In addition to DTI, conventional axial T2-weighted images (echo time [TE]/repetition time [TR]: 95/3500 ms; flip angle: 150°; slice thickness: 5 mm; intersection gap: 1.75 mm; matrix: 448 × 512; field of view [FOV]:
© 2014 The Authors Psychiatry and Clinical Neurosciences © 2014 Japanese Society of Psychiatry and Neurology
362 M. Ota et al.
Psychiatry and Clinical Neurosciences 2015; 69: 360–368
210 × 240 mm2; acquisitions: 1) and fluid attenuation inversion recovery images in the axial plane (TE/ TR: 101/8800 ms; flip angle: 150°; slice thickness: 3 mm; intersection gap: 1.75 mm; matrix: 448 × 512; FOV: 210 × 240 mm2; acquisition: 1) were acquired to exclude cerebrovascular disease. On conventional MRI, no abnormal findings were detected in the brain in any subject included in the study. DTI metrics were calculated by using Tract-Based Spatial Statistics (TBSS) analysis.26 The TBSS algorithm is available in the Functional Magnetic Resonance Imaging of the Brain (FMRIB) Software Library (FSL; version 4.1).27 An FA threshold of 0.20 or higher was set to exclude peripheral tracts.
In addition to the voxel-wise analysis, we examined skeletonized FA and MD data in a regions-ofinterest (ROI) manner for specific white matter tracts derived from the ‘JHU-WhiteMatter-Labels1mm.nii.gz’ regarded as the anatomically standard image atlas in FSL. We evaluated the differences among the two diagnostic groups and healthy subjects using ANCOVA controlling for age, sex, and education, followed by Bonferroni-corrected pairwise comparisons as post-hoc tests using SPSS. A twotailed P-value < 0.05 was regarded as statistically significant.
Statistical analyses We first evaluated the differences in demographic and clinical characteristics among the two diagnostic groups and healthy subjects using ANOVA and between the two patient groups using a two-sample t-test. These tests were performed with SPSS 20 (SPSS Japan, Tokyo, Japan). Skeletonized FA and mean diffusivity (MD) data were analyzed to assess the difference in FA and MD values among the two diagnostic groups and healthy subjects, controlling for age, sex, and education using the Threshold-Free Cluster Enhancement option in FSL’s Randomize with 10 000 permutations. The script uses a permutation-based statistical inference that does not rely on Gaussian distribution of voxels, and runs without having to define an initial clusterforming threshold or carry out a large amount of data smoothing.28,29 The level of statistical significance was set at P < 0.05 with family-wise error (FWE) rate correction for multiple comparisons.
RESULTS Demographic and clinical characteristics of the subjects are shown in Table 1. There was no statistically significant difference in age, sex, or education years across the three groups. There was no significant difference in dose of antidepressant or age of onset between the two patient groups, however the HAM-D score in the mMDD group was significantly higher than that in subjects with aMDD. There were significant decreases of FA in the left superior longitudinal fasciculus (Fig. 1a), left frontooccipital fasciculus (Fig. 1b), and corpus callosum (Fig. 1a,c) in the aMDD patients compared with the healthy subjects (Fig. 1a,c). Significant decreases in FA values were observed diffusely in the mMDD patients compared with the healthy subjects, and the regions exhibiting the FA decreases were similar to those in the subjects with aMDD (Fig. 1d–f). When we compared the FA values between aMDD and mMDD patients, we found no significant difference between the two groups in any region. As for MD, there were significant increases in the left superior
Table 1. Demographic and clinical characteristics of the study sample (mean ± standard deviation)
Male/Female Age (years) Education (years) Age of onset (years) Antidepressant medication (mg)† HAM-D
MDD patients with melancholic features
MDD patients with atypical features
Healthy volunteers
11/10 42.3 ± 10.0 14.7 ± 1.9 33.0 ± 10.8 157.3 ± 186.3 22.0 ± 7.4
11/13 38.3 ± 10.3 14.7 ± 2.2 30.9 ± 9.6 128.7 ± 136.2 14.9 ± 5.2
16/21 40.0 ± 10.0 14.6 ± 2.1
†
Imipramine equivalent dose. HAM-D, Hamilton Rating Scale for Depression; MDD, major depressive disorder.
© 2014 The Authors Psychiatry and Clinical Neurosciences © 2014 Japanese Society of Psychiatry and Neurology
P-value 0.74 0.44 0.93 0.29 0.56
